THE INTEGRATION OF FINANCIAL MARKETS AND GROWTH THE ROLE OF BANKING REGULATION AND SUPERVISION

Transcription

1 Kolegium Gospodarki Światowej Szkoła Główna Handlowa w Warszawie THE INTEGRATION OF FINANCIAL MARKETS AND GROWTH THE ROLE OF BANKING REGULATION AND SUPERVISION 1. Introduction In the latest years many studies focused on the influence of international financial integration on long term economic growth. Despe the abundant empirical research on this topic there is no consensus about the investigated nexus. Respective studies which use various methodologies and datasets point to contradictory results. One of the potential reasons for these discrepancies are thresholds in the financial integration and economic growth nexus concerning the instutional and regulatory qualy 1. Therefore this article addresses the potential influence of regulation as a threshold for the financial integration economic growth nexus. The paper is structured as follows: section two reviews briefly the related strand of lerature. Section three provides the empirical specification and the results. Section four concludes. 1 M. Kose, E. Prasad, A. Taylor, Thresholds in the Process of International Integration, Instute for the Study of Labor (Bonn), IZA Discussion Paper 2009, no

2 2. Lerature review This article borrows from two strands of lerature: one concerned wh the influence of financial integration on economic growth and the second concerned wh regulation and s influence on the financial sector outcomes. The first strand of related lerature 2 identifies many possible channels through which financial integration can influence economic growth. First of all financial integration enables the fulfillment of the financial systems functions internationally. surpluses can be invested more efficiently as the effects of information and transaction costs are migated. integration enables also international risk sharing 3. On the other hand financial integration has a negative influence on the extent of domestic savings as creates more profable allocation possibilies of these surpluses. Second of all financial integration affects the real economy also through indirect channels. It contributes to the development of domestic financial markets and stimulates the volume of international trade. Both factors are economic growth enhancing. Nevertheless increasing financial integration creates the risk of economic volatily and transmission of crises. The functioning of the mentioned transmission channels is influenced by thresholds concerning the instutional and regulatory qualy 4. An important threshold affecting the influence of financial integration on economic growth may be regulation and supervision. The lack of prudential supervision may be and impediment to the allocation channel, may lead to overborrowing and to boom and bust cycles 5. Although regulation seems to be an important threshold to the financial integration and economic growth nexus this two issues have not been analyzed jointly in the lerature so far. Therefore this study is based also on a second strand of lerature concerned wh various features of regulation and supervision as factors influencing financial stabily 6.This article especially builds on 2 J. Babecký, L. Komárek, Z. Komárková, Integration of Stock Markets among New EU Member States and the Euro Area, Warwick Economic Research Papers 2010, no. 849; A. Bonfiglioli, Integration, Productivy and Capal Accumulation, Journal of International Economics 2008, vol. 76 (2), pp ; M. Osada, S. Sao, Integration and Economic Growth: An Empirical Analysis Using International Panel Data from , Paper for the third annual workshop of the BIS Asian Research Networks held on March 26, 2010, 3 M. Kose, E. Prasad, M. Terrones, How Does Globalization Affect Risk Sharing? Patterns and Channels, International Monetary Fund (Washington), Working Papers 2007, no. 07/ E. Glaeser, R. La Porta, F. Lopez de Silanes, A. Schleifer, Do instutions cause growth?, Journal of Economic Growth 2004, vol. 9 (3), pp N. Cetorelli, L. Goldberg, Global Banks and International Shock Transmission: Evidence from the Crisis, IMF Economic Review 2011, vol. 59 (1), pp R. Levine, S. Zervos, Stock Markets, Banks, and Economic Growth, The American Economic Review 1998, vol. 88 (3), pp ; N. Loayza, R. Ranciere, Development, Fragily, and Growth, Journal of Money, Cred, and 2006, vol. 38 (4), pp ; A. Demirgüc Kunt, E. Detragiache, 130

3 The integration of financial markets and growth the role of regulation and supervision the paper of Barth, Caprio and Levine (2004), which investigates the relationship between a broad array of regulatory and supervisory practices and sector efficiency and fragily. 3. Methodology and the results Following the line of an extensive strand of related lerature this study uses as financial integration measure the annual stock of foreign assets and liabilies to GDP provided by Lane and Milesi Ferretti (2008). The macroeconomic data is taken from PWT 7.0 and the World Bank database. The study uses also the average years of schooling as a proxy for human capal- a measure provided by Barro and Lee (2000) and an indicator of instutional qualy provided by Kaufmann, Kraay and Mastruzzi (2010) 7. The study covers the period of and a sample of 69 countries. To investigate the role of regulation and supervision the study uses a broad set of indicators computed on the base of the paper by Barth, Caprio and Levine (2004) 8 and the latest update of the World Bank Regulatory Database (2008). The latter encompasses indicators of: regulatory restrictions on bank activies and the mixing of and commerce, regulations on domestic and foreign bank entry, regulations on capal adequacy, depos insurance system regulation, supervisory power, loan classification stringency, provisioning standards, and diversification guidelines, regulations fostering information disclosure and private sector monoring of banks and government ownership. The study uses the respective detailed regulatory elements and the broad regulation as a whole as indicators. The measures of supervision, entry into, activy, capal requirements, private monoring and depos insurance indicate regulation stringency. Higher indicators stand for greater stringency. In the case of foreign bank ratio, government owned bank ratio and concentration pure numbers are computed. Following Barth, Caprio and Levine (2004) the paper uses the absolute and the first principal component versions of the indicators. The reported results are principal component versions but the exercises are performed for absolute versions as well. Table 1 presents the correlation coefficients between the respective forms of sector features. Basel Core Principles and bank soundness. Does compliance matter?, Journal of Stabily 2011, vol. 7 (4), pp D. Kaufmann, A. Kraay, M. Mastruzzi, The Worldwide Governance Indicators, Methodology and Analytical Issues, World Bank (Washington), Policy Research Working Paper J. Barth, G. Caprio, R. Levine, Bank regulation and supervision: what works best?, Journal of Intermediation 2004, vol. 13, pp

4 Table 1. Correlation among sector features supervision Foreign banks Government banks concentration supervision Foreign banks 0.19* 0.25*** 1 Government banks *** concentration Entry into *** 0.05 activy Capal requirements 0.27*** Private monoring 0.19* Depos insurance Continuation of Table 1 Entry into activy Capal requirements Private monoring Depos insurance Entry into 1 1 activy Capal requirements * 1 Private monoring 0.17* Depos insurance * significance at 0,1 level, ** significance at 0,05 level, *** significance at level Source: authors coputations based on: J. Barth, G. Caprio, R. Levine, Bank regulation and supervision: what works best?, Journal of Intermediation 2004, vol. 13, pp and the World Bank Regulatory Database (2008). Summary statistics indicate that there is significant posive correlation between supervision and the ratio of government owned banks. In contrast- supervision is significantly negatively correlated wh capal requirement stringency and private monoring restrictiveness. The ratio of foreign owned banks is posively correlated wh concentration. This may be due the fact that foreign banks are owned by a limed number of international investors. One can glean from the table that entry into stringency is negatively correlated wh the ratio of government owned banks. The entry into stringency is also negatively correlated 132

5 The integration of financial markets and growth the role of regulation and supervision wh the depos insurance restrictiveness. There is also significant negative correlation between capal requirements stringency and the extent of private monoring. To investigate the impact of financial integration on growth this paper builds on the augmented neoclassical growth model 9. The empirical specification closely follows the line of the mentioned related studies. The specification is based on dynamic panel data estimation. The basic model has the following form: Δy 8 = α y α ΔPOP + α INV α FIBR + ξ + η + ε. i 2 t 3 + α HC 4 + α FI 5 + α MACRO 6 + α BR 7 + (1) To eliminate the influence of short term cyclical fluctuations and to account for market structure adjustments the estimation is based on five year averages of the underlying data. Growth rates (Δy ) are defined as the log difference of real GDP per capa, y 0 denotes the log of real GDP per capa five years prior to period t and accounts for the convergence effect. ΔPOP stands for the log difference of population, INV denotes the gross fixed capal formation ratio to GDP, HC stands for the average years of schooling as a proxy for human capal, FI signifies the financial integration measure. MACRO is a set of control variables which are robust correlates of growth. BR signifies the respective regulatory variables FIBR stands for the interactions between financial integration and regulation. The introduction of the interactive variables allows to investigate how regulatory variables affect the marginal effect of financial openness on growth. To check for a potentially optimal level of regulation squared values of the variables are included in the second step of the study. Subsequently the basic model is modified as follows: the basic model is modified as follows: 2 Δy = α1y 0 + α 2ΔPOP + α3 INV + α 4HC + α5 FI + α 6MACRO + α 7 BR + 2 (2) α8 FIBR + ξ i + η t + ε. Due to potential endogeney and because of the inclusion of time invar Due to potential endogeney and because of the inclusion of time invariant variables in the regression the applied estimation technique is the Blundell Bond system variables in the regression the applied estimation technique is the Blundell-B GMM. The Hausman test points to the right selection of the estimator, the Sargan test does not allow to reject the null hypothesis of the validy of instruments. For what concerns the stationary of the data the Fisher type ADF test wh one lag shows that in the case of all variables one can reject the null hypothesis that all panels contain a un root in favour of the alternative hypothesis that at least one panel is stationary. In the first step of the study a set of ten regressions conforming to the baseline model (1) is estimated. The main results are shown in Table 2 and 3. In the second step of the study the estimation is performed by means of 5 regressions conforming 9 G. Mankiw, D. Romer, D. Weil, A Contribution to the Empirics of Economic Growth, The Quarterly Journal of Economics 1992, vol. 107 (2), pp

6 to the modified model (2) wh squared interactions. In the third step robustness checks for subsamples are performed. The basic sample is divided into industrialized, emerging and developing economies. Table 2. Baseline regression results Inial log real GDP per capa 0.06** 0.02** 0.02** * 0.02** Population growth Average years of schooling Gross fixed capal formation 10.61*** 10.60*** 10.59*** 10.61*** 10.92*** 0.06*** 0.05*** ** *** *** *** *** *** Openness 10( 4) Inflation 2*10 ( 4) *** 2*10 ( 4) *** 2*10 ( 4) *** 2*10 ( 4) *** 2*10 ( 4) *** Instutional qualy integration depth Broad regulation supervision Foreign banks ratio Government owned banks ratio concentration Interaction wh broad regulation 0.06 *** 0.05*** ** ** 10 ( 4) 10 ( 4) 2*10 ( 4) 10 ( 4) 3*10 ( 4) * *10 ( 3) 53 41**

7 The integration of financial markets and growth the role of regulation and supervision Table 3. Baseline regression results (cont.) Inial log real GDP per capa Population growth Average years of schooling Gross fixed capal formation *** 10.56*** 10.58*** 10.49*** 10.55*** 10.49*** 0.05*** 0.05*** 0.06*** 0.05*** 0.05*** *** *** *** *** *** Openness Inflation 2*10 ( 4) *** 2*10 ( 4) *** 2*10 ( 4) *** 2*10 ( 4) *** 2*10 ( 4) *** Instutional qualy integration depth Entry into activy Capal requirements Private monoring Depos insurance 0.02 *** *** 0.05*** 0.05*** *** ** Interaction wh broad regulation 135

8 Table 4. The results of the regression wh quadratic interaction Inial log real GDP per capa 0.21*** 0.21*** 0.21*** 0.21*** 0.21*** 0.21*** Population growth Average years of schooling Gross fixed capal formation 1*** 10.09*** 10.06*** 10.05*** 10.07*** 10.05*** 0.04*** 0.04*** 0.04*** 0.04*** 0.04*** 0.04*** *** *** *** *** *** *** Openness Inflation Instutional qualy integration depth Broad regulation square supervision square Entry into square activy square Capal requirements square Depos insurance square

9 The integration of financial markets and growth the role of regulation and supervision Interaction term square 2*10 ( 4) *** 10( 4) Table 5. Robustness checks for subsamples Inial log real GDP per capa Population growth Average years of schooling Gross fixed capal formation 0.32*** 0.32*** *** 10.04*** 0.03** 0.05 *** Openness Inflation 10 ( 4) * Instutional qualy integration 0.02 depth ** Broad regulation supervision ** 0.02* 0.28*** 0.35*** *** 0.87* 0.03** 0.05 *** 10 ( 4) * ** * 0.16** 0.36*** * 10.84*** *** *** ** *** Foreign banks *** *** *** *** *** *** 0.20*** 0.36*** *** 10.73*** * *** *** *** 137

10 Government owned banks concentration 0.03 Interaction term Inial log real GDP per capa 0.30*** 0.35*** *** 0.36*** *** 0.35*** *** 0.34*** *** 0.35*** 0.08 Population growth *** 0.90* *** 0.89* *** 0.84* *** 0.85* 0.66* 10.70*** 0.87* Average years of schooling 0.03** ** * ** ** 0.06 Gross fixed capal formation *** *** 1*** *** Openness Inflation 10 ( 4) * 10 ( 4) * 10 ( 4) * 10 ( 4) * 10 ( 4) * Instutional qualy integration * 0.03* 0.03* depth 5*10 ( 4) * 5*10 ( 4) * 4*10 ( 4) * 6*10 ( 4) ** Entry into regulation

11 The integration of financial markets and growth the role of regulation and supervision activy regulation Capal requirements Private monoring Depos insurance Interaction term * The estimates in the respective rows are computed for industrialized, emerging and developing countries. 0.76*** Conclusions and further research The results of the baseline regression are presented in tables 2 and 3. The exercise confirms the results of the majory of related studies concerning the negative relation between financial integration and long term growth. This effect is significant for nine out of ten regressions. The coefficients of the control variables are in line wh those obtained in similar papers. As far as the regulatory measures are concerned the exercise points to a posive and significant impact of overall regulation restrictiveness on growth. The estimation points moreover to a significant posive effect of capal requirements stringency on growth. In contrast- the influence of supervision restrictiveness on growth is negative and significant. Other regulatory features seem not to matter for growth, neher do general sector features like foreign banks ratio, government owned banks ratio or concentration. The coefficient of the interactive term is posive in all regressions and significant in two out of ten panels. This result might imply that strict regulation turns the negative impact of financial integration on growth posive. Table 4 presents the results for the regression wh the quadratic interaction between financial integration and regulation. The sign of the coefficient of the financial integration variable is again negative. The coefficients of the squares of the respective regulatory variables are posive yet insignificant. The quadratic interactive term is posive as well. This confirms the result obtained in the first step of the study that restrictive regulation may turn the negative impact of financial 139

12 integration on growth into posive. The result may indicate moreover that there is an optimal level of regulation, although this effect is statistically insignificant. The tables 5 and 6 provide the robustness checks for subsamples. The results differ depending on the group of countries investigated. integration seems to have a posive insignificant effect on growth in industrialized countries, a negative insignificant impact in emerging countries and a negative significant influence in developing countries. Similarly varies the importance of regulatory features. Restrictive regulation contributes to growth significantly and posively in emerging countries while significantly and negatively in developing countries. Its impact on long term growth in industrialized countries is posive and insignificant. On the other hand supervision has a negative significant coefficient for the industrialized countries sample while remains posive and insignificant for emerging and developing economies. The industrialized countries are also characterized by a negative significant relation between capal requirements stringency and depos insurance extent and growth. In emerging economies this impact is negative and insignificant in both cases, for the developing countries stringent capal requirements have a posive insignificant impact on growth while depos insurance a negative insignificant effect. One can also glean from table 5 and 6 that the results concerning the interactive term obtained in the first step of he study are only robust for the developing country sample. Here the posive coefficient of the interactive variable may suggest that stringent regulation migates the negative influence of financial integration on growth in those countries. In emerging economies stringent regulation seems not to affect the negative impact of financial integration on growth since the coefficient is negative. For the industrialized country sample no robust conclusions can be drawn about the role of the interactive term. The analyzed topic surely needs further research. The role of regulation in shaping the real effects of financial markets integration has gained on importance especially after the latest global financial crisis from A serious limation to the study is the constraint that the data on regulation is cross sectional and does not reflect the recent changes in the regulatory framework. Nevertheless the results of the study may be a point of departure for further research on the emerging discussion about the role of regulation in the financial integration- growth nexus. 140

13 The integration of financial markets and growth the role of regulation and supervision References 1. Babecký J., Komárek L., Komárková Z., Integration of Stock Markets among New EU Member States and the Euro Area, Warwick Economic Research Papers 2010, no Barro R., Lee W., International Data on Educational Attainment: Updates and Implications, CID Working Paper 2000, no. 42, working papers/cid working paper no Barth J., Caprio G., Levine R., Bank regulation and supervision: what works best?, Journal of Intermediation 2004, vol. 13, pp Bonfiglioli A., Integration, Productivy and Capal Accumulation, Journal of International Economics 2008, vol. 76 (2), pp Cetorelli N., Goldberg L., Global Banks and International Shock Transmission: Evidence from the Crisis, IMF Economic Review 2011, vol. 59 (1), pp Demirgüc Kunt A., Detragiache E., Basel Core Principles and bank soundness. Does compliance matter?, Journal of Stabily 2011, vol. 7 (4), pp Glaeser E., La Porta R., Lopez de Silanes F., Schleifer A., Do instutions cause growth?, Journal of Economic Growth 2004, vol. 9 (3), pp Kaufmann D., Kraay A., Mastruzzi M., The Worldwide Governance Indicators, Methodology and Analytical Issues, World Bank (Washington), Policy Research Working Paper Kose M., Prasad E., Terrones M., How Does Globalization Affect Risk Sharing? Patterns and Channels, International Monetary Fund (Washington), Working Papers 2007, no. 07/ Kose M., Prasad E., Taylor A., Thresholds in the Process of International Integration, Instute for the Study of Labor (Bonn), IZA Discussion Paper 2009, no Levine R., Zervos S., Stock Markets, Banks, and Economic Growth, The American Economic Review 1998, vol. 88 (3), pp Loayza N., Ranciere R., Development, Fragily, and Growth, Journal of Money, Cred, and 2006, vol. 38 (4), pp Mankiw G., Romer D., Weil D., A Contribution to the Empirics of Economic Growth, The Quarterly Journal of Economics 1992, vol. 107 (2.), pp Osada M., Sao S., Integration and Economic Growth: An Empirical Analysis Using International Panel Data from , Paper for the third annual workshop of the BIS Asian Research Networks held on March 26, 2010, 141

14 Summary The integration of international financial markets and growth the role of regulation and supervision The study is aimed at investigating the role of regulation and supervision in the relation between financial integration and long term growth by means of dynamic panel model estimation. The study covers a sample of 69 countries during the period The study gives some new results on the impact of regulation on the investigated relation. regulation can turn the negative impact of financial integration on growth into posive. Moreover the results indicate that the respective regulatory features play various roles in shaping the effects of financial integration on growth. Keywords: financial integration, regulation, supervision, growth JEL classification: F36, G21, C23